DESCRIPTION: (Adapted from Applicant's Abstact). A large body of epidemiologic evidence suggests that estrogen is involved in the aetiology of breast cancer, but variables used a surrogates of estrogen levels, such as the age at menarche, are only weakly related to risk. Measured levels of endogenous estrogens also have generally been weakly related to risk, and it is not clear whether a single or even several measurements, over time, would be a good measure of cumulative estrogen exposure. Estrogens are strong determinants of BMD, which may, then, serve as a good measure of cumulative exposure. Recent evidence from studies of older women suggests that BMD may predict the risk of breast cancer. For example, in the Framingham Study, a relatively crude measure of bone mass from radiography was strongly related to the incidence of breast cancer; the rate ratio for the comparison of the highest to the lowest quartile of bone mass was 4.5. This application proposes to study the relation of BMD of the hip (neck of the femur) and the lumbar spine to the risk of breast cancer in a case-control study of breast cancer in relation to contraceptive use, currently being conducted in Cape Town, South Africa, among non-white women under the age of 55 years. In this study, extensive information is collected on risk factors for breast cancer. Cases and controls will be invited to have BMD measurements and will also be questioned about physical activity on the job, at home, and at leisure. The relative risk will be estimated for women with higher levels of BMD relative to women in the lowest quartile with control for important breast cancer and BMD risk factors. Female hormone supplements are rarely used in this population and will not be confounders. Since the subject population is at low risk of breast cancer, a relationship between BMD and risk may show up more strongly than in higher-risk populations. If BMD at younger ages is a predictor of breast cancer risk, this would be useful for the identification of women at high risk at the ages at which American women often have BMD measured (i.e., at the time of menopause).